Chinese researchers have developed an advanced artificial intelligence model (AI) capable of detecting early gastric cancer using tomographies not with contrast (TC) without contrast.
This is reported by Xinhua News Agency, a TV BRICS partner.
The AI system is called UVA (Gastric Cancer Risk Assessment Procedure with artificial intelligence).
Also read: Osun: Nigeria, Chinese partner, will build agroindustrial assets in Iragbiji
It allows the detection of stomach tumors in the early stages, trusting only in non -invasive TC images, widely available, previously considered ineffective for the diagnosis of gastric cancer.
The model was trained in almost 100,000 computerized tomographs of 20 medical institutions in China.
It demonstrated a sensitivity of 85.1 percent in the identification of early gastric tumors, overcoming the performance of experienced human radiologists.
The AI uses a new image registration technique, aligning CT data improved by contrast with explorations without contrast, allowing the model to identify tumors with high precision.
Clinical trials made in two hospitals showed that the model increased detection rates to 24.5 percent, including cases in asymptomatic individuals.
In a retrospective study, the grape identified signs of cancer in computerized tomographs taken two to ten months before formal diagnosis.
The model has already been deployed in hospitals in the provinces of Zhejiang and Anhui, with a greater implementation planned both in China and abroad.
The broader strategic impulse of China in the medical diagnosis promoted by AI has thrown several advances nationwide.
The researchers also recently developed an artificial intelligence tool for the detection of cervical cancer, achieving precision along or superior to human experts, said the source..